Although infection with Mycobacterium tuberculosis (MTb) is the largest cause of death in AIDS, and it is the major cause of HIV-associated pulmonary disease globally, little is known about how TB co-infection impacts long term immune reconstitution or the HIV-1 reservoir. The Cambodian Early vs. Late Introduction of Anti- retrovirals (ART) (CAMELIA) randomized clinical trial showed that early initiation of ART (2 weeks after TB therapy initiation versus late ART at 8 weeks), resulted in a significant (34%) decrease in mortality in TB+/HIV+ patients with a CD4+ T cell count <200 cells/mm, a survival benefit that persisted for at least 3 years. Our preliminary data examining T cell responses in TB+/HIV+ individuals from the CAMELIA randomized clinical trial in a nested substudy, demonstrated that TB and TB-associated immune reconstitution syndrome (TB-IRIS) results in profound changes in the T cell compartment for at least 8 months after treatment initiation. For example, TB co-infection and TB-IRIS is associated with massive T cell activation including elevated CCR5+ T cells and results in significant impact on the proportions of central memory (TCM) (CD62L+CD45RA-) CD4+T cells, the major peripheral HIV reservoir component and effector memory. In patients who experienced TB- IRIS, we observed a reciprocal expansion of T effector (TEM) (CD62L-CD45RA-) CD4+T cells, which are associated with control TB. Furthermore, TB-IRIS patients had elevated CXCR3+CCR6+CD4+ T cells, which are thought to mediate a large portion of CD4+ anti-TB responses, and these CXCR3+CCR6+CD4+ T cells remained significantly elevated for at least 8 months after the initiation of ART. Based on our preliminary data, we hypothesize that the distinct cellular responses we observed in TB-IRIS may represent important protective responses to TB. We hypothesize that evaluating these responses in former CAMELIA patients cured of TB and on stable ART in a TB endemic environment will lead to identification of novel antigens important for protective responses and future vaccine development. Host biomarkers of TB infection, TB-IRIS and TB/HIV co-infection and linked clinical outcomes are lacking. Using RNA microarrays, we will determine signatures associated with TB co-infection, TB-IRIS, and antigen specific TB responsiveness. We hypothesize that this analysis in this extremely well-characterized patient cohort will provide robust and novel biomarkers of TB co- infection, containment of TB disease, and protective responses associated with in vitro assays. Finally, using archived PBMC, from TB+/ HIV+ patients from CAMELIA or TB-/HIV+ patients with similar baseline characteristics who began ART at the same time, we will test the hypothesis that TB co-infection and TB-IRIS modulate HIV reservoir tropism after ART initiation the months after ART initiation and years later and potentially also influence reservoir size. Findings from these experiments will be of great value for improving both treatment of TB/HIV co-infection and our general understanding of the host immune response to TB/HIV as well as providing the first evidence indicating that TB co-infection can affect the composition of the HIV reservoir.